Skip to main content

Engagement Tailoring Record -- AI Support Copilot

Engagement: AI Support Copilot Pilot Owner: Amit (POD Lead) + Shivani (PM) Version: 1.0 Date: 2026-05-01 Framework ref: Doc 01, Section 5.1

The Gyde POD Operation Framework is designed to be tailored per engagement. This document records every deviation from the default framework practices, with rationale. Non-negotiable items (Doc 01, Section 5.1) cannot be tailored away without Engineering Leadership approval.


1. Non-Negotiable Compliance

All five non-negotiables are upheld in this engagement. None have been waived.

#Non-NegotiableStatusHow Fulfilled
1Threat modeling and secrets managementCompliantThreat Model (#10) produced; secrets via env vars / GCP Secret Manager
2Evaluation before productionCompliantEvaluation Plan (#6), Eval Harness (#20), Golden Dataset (#21), 1,000-question synthetic eval
3Versioned data and promptsCompliantAll prompts, datasets, configs in Git; every change tracked
4Audit trail for AI decisionsCompliantEvery copilot decision logged to MongoDB audit_log with full pipeline output
5Incident response readinessCompliantRunbooks included in Knowledge Transfer Package (#34)

2. Tailoring Decisions

2.1 Engagement Lifecycle (Doc 03, Section 2)

DefaultTailored ToRationaleImpact
5 phases: Discovery → Sprint Zero → Build → Hardening → Handover3 phases: Discovery → Sprint 1 (Build) → Sprint 2 (Build + Hardening + Handover)16-day pilot timeline; Sprint Zero activities absorbed into Discovery and Sprint 1 Day 1; Hardening and Handover compressed into Sprint 2Higher risk; mitigated by pre-approved scope cuts in Risk Register
M0 through M6 milestonesM0 through M3 onlyPilot scope; M4-M6 (production deployment, stable operation, engagement close) are out of scopeClient understands pilot != production
Discovery: 1-3 weeksDiscovery: 1 day (May 1)Internal simulation; team already aligned; Discovery call answers captured same dayAcceptable for simulation; real client engagement would take longer

2.2 Sprint Cadence (Doc 04, Section 2)

DefaultTailored ToRationaleImpact
2-week (10-day) sprintsSprint 1: 10 days, Sprint 2: 6 daysHard deadline May 16; Sprint 2 is compressed but covers only hardening + handover scopeSprint 2 has less buffer; contingency plan addresses this
Backlog refinement on Day 9Sprint 1 only; Sprint 2 too short for separate refinementSprint 2 backlog is fully planned during Discovery; no refinement ceremony neededAcceptable -- Sprint 2 scope is fixed
WIP limit: 1 in-progress story per memberRelaxed to 2 for Sprint 1Parallel tracks (infra + data + pipeline) require some members to context-switchMonitor in retro; revert if quality suffers

2.3 QA Practices (Doc 04, Section 6)

DefaultTailored ToRationaleImpact
QA buddy assigned per story; rotates each sprintQA buddy assigned per story; limited rotation due to 2-sprint engagementOnly 2 sprints; full rotation not achievable; buddy assignments prioritize domain knowledgeAcceptable for pilot; some members won't QA-buddy every part of the system
Adversarial eval run every 2 sprintsRun once in Sprint 2Only 2 sprints total; adversarial set created and run in Sprint 2Compliant -- adversarial eval happens before delivery

2.4 Security (Doc 14)

DefaultTailored ToRationaleImpact
Full authentication and authorizationNo authentication in pilotInternal pilot with simulated data; single-user accessProduction must add JWT auth / SSO; documented in productionization note
Data processing agreements for LLM providersDeferred to productionPilot uses simulated data; no real customer PIIProduction must evaluate DPA with Google or switch to on-prem LLM
Network-level security reviewDeferred to productionPilot runs on internal GCP VM; no external traffic in pilotProduction requires firewall rules, VPN, TLS

2.5 Documentation (Doc 19)

DefaultTailored ToRationaleImpact
Separate Architecture Sketch + final Architecture DocumentArchitecture Sketch (#5) serves as architecture doc; updated to final during Sprint 2 (#31)Pilot is short; maintaining two documents for same content is overheadSingle document evolves from sketch to final
Monthly steering committee packNot applicableEngagement is 16 days; no monthly cadence existsReplaced by weekly status email + sprint demos

2.6 Operations (Doc 18)

DefaultTailored ToRationaleImpact
Monitoring dashboardsNot built in pilotNo production traffic; monitoring adds overhead without valueDocumented as production requirement in productionization note
On-call rotationNot applicableNo production deployment; pilot is demo-onlyDocumented in KT package as production requirement
Rollback planNot applicableNo production deploymentDocumented in KT package

2.7 MLOps (Doc 12)

DefaultTailored ToRationaleImpact
Model version registryPrompt versions tracked in GitNo model training / fine-tuning; RAG architecture uses foundation model as-isAdequate for pilot; production may need formal registry if fine-tuning is added
Drift detectionNot applicableNo production traffic to detect drift againstDocumented as production requirement
A/B testing frameworkNot applicableSingle pipeline variant; no traffic to splitCould be useful in production for prompt iteration

3. Items Explicitly NOT Tailored

These framework practices are followed as-is, without modification:

PracticeFramework RefWhy Kept
Threat modelingDoc 14Non-negotiable; initial model produced at Discovery, full review at Sprint 2
Evaluation harness with golden datasetDoc 17Non-negotiable; eval gates every release
Versioned prompts and dataDoc 10Non-negotiable; all prompts in Git as template files
Audit loggingDoc 01Non-negotiable; every copilot decision logged
Sprint demos to clientDoc 04Essential for client alignment; both demos happening
Sprint retroDoc 04Essential for team improvement; both retros happening
Risk register (living)Doc 03Updated weekly; visible to client
Definition of DoneDoc 04Full DoD applied to every story including QA buddy verification
Change request processDoc 03Active for any scope/schedule changes
Decision logging (ADRs)Doc 19Key decisions documented as they occur

4. Approval

RoleNameApprovalDate
POD LeadAmit____________________
Implementation ManagerShivani____________________

Note: No Engineering Leadership approval required because all five non-negotiables are upheld. Tailoring decisions affect only engagement-specific practices, not framework principles.


Change Log

DateChangeBy
2026-05-01Initial tailoring record createdAmit + Shivani

This record ensures transparency about what we changed and why. Future engagements can reference this as a precedent for similar pilot-scoped tailoring.